The atmospheric sulfur cycle in ECHAM-4 and its impact on the shortwave radiation

 The atmospheric general circulation model ECHAM-4 is coupled to a chemistry model to calculate sulfate mass distribution and the radiative forcing due to sulfate aerosol particles. The model simulates the main components of the hydrological cycle and, hence, it allows an explicit treatment of cloud transformation processes and precipitation scavenging. Two experiments are performed, one with pre-industrial and one with present-day sulfur emissions. In the pre-industrial emission scenario SO₂ is oxidized faster to sulfate and the in-cloud oxidation via the reaction with ozone is more important than in the present-day scenario. The atmospheric sulfate mass due to anthropogenic emissions is estimated as 0.38 Tg sulfur. The radiative forcing due to anthropogenic sulfate aerosols is calculated diagnostically. The backscattering of shortwave radiation (direct effect) as well as the impact of sulfate aerosols on the cloud albedo (indirect effect) is estimated. The model predicts a direct forcing of −0.35 W m^-2 and an indirect forcing of −0.76 W m^-2. Over the continents of the Northern Hemisphere the direct forcing amounts to −0.64 W m^-2. The geographical distribution of the direct and indirect effect is very different. Whereas the direct forcing is strongest over highly polluted continental regions, the indirect forcing over sea exceeds that over land. It is shown that forcing estimates based on monthly averages rather than on instantaneous sulfate pattern overestimate the indirect effect but have little effect on the direct forcing. Received: 16 October 1996/Accepted: 24 October 1996     

A time-slice experiment with the ECHAM4 AGCM at high resolution: the impact of horizontal resolution on annual mean climate change

 The climate response to increasing levels of atmospheric greenhouse gases, prescribed according to the International Panel of Climate Change (IPCC) scenario IS92a, is studied in two model simulations. The reference simulation is a transient response experiment performed with a medium-resolution (T42) coupled general circulation model of the atmosphere and ocean (ECHAM4/OPYC) developed at the Max-Planck-Institute for Meteorology. For two 30-year “time slices”, representing the present-day climate and the future climate at the time of effective CO₂ doubling, the annual mean climate states are compared with those obtained from the high-resolution (T106) ECHAM4 model forced with monthly sea surface temperatures and sea-ice from the coupled model. The large-scale changes in temperature, zonal wind, sea-level pressure and precipitation are broadly similar. This applies, in particular, to the respective zonal means. In general, except for precipitation, the responses in the time-slice experiments are slightly weaker than those simulated in the coupled model due to a smaller effect of the horizontal resolution on the simulations of the future (warmer) period than on the simulations of the present period. On a regional scale, the impact of horizontal resolution is smaller in the Southern than in the Northern Hemisphere, where the response differences are caused mainly by changes in the positions of the stationary waves. Although the precipitation responses are broadly similar, there are few notable exceptions such as a more pronounced maximum over the equatorial oceans in the T106 experiment but a weaker response over low-latitude land areas. Differences in precipitation response are found especially in areas with strong topographical control such as South America, for example. Received: 17 January 2000 / Accepted: 7 July 2000     

A nonlinear impulse response model of the coupled carbon cycle-climate system (NICCS)

 Impulse-response-function (IRF) models are designed for applications requiring a large number of climate change simulations, such as multi-scenario climate impact studies or cost-benefit integrated-assessment studies. The models apply linear response theory to reproduce the characteristics of the climate response to external forcing computed with sophisticated state-of-the-art climate models like general circulation models of the physical ocean-atmosphere system and three-dimensional oceanic-plus-terrestrial carbon cycle models. Although highly computer efficient, IRF models are nonetheless capable of reproducing the full set of climate-change information generated by the complex models against which they are calibrated. While limited in principle to the linear response regime (less than about 3 ^∘C global-mean temperature change), the applicability of the IRF model presented has been extended into the nonlinear domain through explicit treatment of the climate system's dominant nonlinearities: CO₂ chemistry in ocean water, CO₂ fertilization of land biota, and sublinear radiative forcing. The resultant nonlinear impulse-response model of the coupled carbon cycle-climate system (NICCS) computes the temporal evolution of spatial patterns of climate change for four climate variables of particular relevance for climate impact studies: near-surface temperature, cloud cover, precipitation, and sea level. The space-time response characteristics of the model are derived from an EOF analysis of a transient 850-year greenhouse warming simulation with the Hamburg atmosphere-ocean general circulation model ECHAM3-LSG and a similar response experiment with the Hamburg carbon cycle model HAMOCC. The model is applied to two long-term CO₂ emission scenarios, demonstrating that the use of all currently estimated fossil fuel resources would carry the Earth's climate far beyond the range of climate change for which reliable quantitative predictions are possible today, and that even a freezing of emissions to present-day levels would cause a major global warming in the long term. Received: 28 January 2000 / Accepted: 9 March 2001     

The effect of snow on Antarctic sea ice simulations in a coupled atmosphere-sea ice model

 The effect of a snow cover on sea ice accretion and ablation is estimated based on the ‘zero-layer’ version sea ice model of Semtner, and is examined using a coupled atmosphere-sea ice model including feedbacks and ice dynamics effects. When snow is disregarded in the coupled model the averaged Antarctic sea ice becomes thicker. When only half of the snowfall predicted by the atmospheric model is allowed to land on the ice surface sea ice gets thicker in most of the Weddell and Ross Seas but thinner in East Antarctic in winter, with the average slightly thicker. When twice as much snowfall as predicted by the atmospheric model is assumed to land on the ice surface sea ice also gets much thicker due to the large increase of snow-ice formation. These results indicate the importance of the correct simulation of the snow cover over sea ice and snow-ice formation in the Antarctic. Our results also illustrate the complex feedback effects of the snow cover in global climate models. In this study we have also tested the use of a mean value of 0.16 Wm^-1 K^-1 instead of 0.31 for the thermal conductivity of snow in the coupled model, based on the most recent observations in the eastern Antarctic and Bellingshausen and Amundsen Seas, and have found that the sea ice distribution changes greatly, with the ice becoming much thinner by about 0.2 m in the Antarctic and about 0.4 m in the Arctic on average. This implies that the magnitude of the thermal conductivity of snow is of considerable importance for the simulation of the sea ice distribution. An appropriate value of the thermal conductivity of snow is as crucial as the depth of the snow layer and the snowfall rate in a sea ice model. The coupled climate models require accurate values of the effective thermal conductivity of snow from observations for validating the simulated sea ice distribution under the present climate conditions. Received: 20 November 1997/Accepted: 27 July 1998     

The modification of greenhouse gas warming by the direct effect of sulphate aerosols

 The Canadian Centre for Climate Modelling and Analysis (CCCma) second generation climate model (GCMII) consists of an atmospheric GCM coupled to mixed layer ocean. It is used to investigate the climate response to a doubling of the CO₂ concentration together with the direct effect of scattering by sulphate aerosols. As expected, the aerosols offset some of the greenhouse gas (GHG) warming; the global annual mean screen temperature change due to doubled CO₂ is 3.4 °C in this model and this is reduced to 2.7 °C when an estimate of the direct effect of anthropogenic sulphate aerosols is included. The pattern of climate response to the comparatively localized aerosol forcing is not itself localized, and it bears a striking resemblance to the response pattern that arises from the globally distributed change in GHG forcing. This “non-local” response to “localized” forcing indicates that the pattern of climate response is determined, to first order, by the overall magnitude of the change in forcing rather than its detailed nature or structure. Feedback processes operating in the system apparently determine this pattern by locally amplifying and suppressing the response to the magnitude of the change in forcing. The influence of the location of the change in forcing is relatively small. These “non-local” and “local” effects of aerosol forcing are characterized and displayed and some of their consequences discussed. Effects on the moisture budget and on the energetics of the global climate are also examined. Received: 10 June 1997 / Accepted: 8 January 1998     

Glacial termination: sensitivity to orbital and CO2 forcing in a coupled climate system model

 To study glacial termination and related feedback mechanisms, a continental ice dynamics model is globally and asynchronously coupled to a physical climate (atmosphere-ocean-sea ice) model. The model performs well under present-day, 11 kaBP (thousand years before present) and 21 kaBP perpetual forcing. To address the ice-sheet response under the effects of both perpetual orbital and CO₂ forcing, sensitivity experiments are conducted with two different orbital configurations (11 kaBP and 21 kaBP) and two different atmospheric CO₂ concentrations (200 ppmv and 280 ppmv). This study reveals that, although both orbital and CO₂ forcing have an impact on ice-sheet maintenance and deglacial processes, and although neither acting alone is sufficient to lead to complete deglaciation, orbital forcing seems to be more important. The CO₂ forcing has a large impact on climate, not uniformly or zonally over the globe, but concentrated over the continents adjacent to the North Atlantic. The effect of increased CO₂ (from 200 ppmv to 280 ppmv) on surface air temperature has its peak there in winter associated with a reduction in sea-ice extent in the northern North Atlantic. These changes are accompanied by an enhancement in the intensity of the meridional overturning and poleward ocean heat transport in the North Atlantic. On the other hand, the effect of orbital forcing (from 21 kaBP to 11 kaBP) has its peak in summer. Since the summer temperature, rather than winter temperature, is found to be dominant for the ice-sheet mass balance, orbital forcing has a larger effect than CO₂ forcing in deglaciation. Warm winter sea surface temperature arising from increased CO₂ during the deglaciation contributes to ice-sheet nourishment (negative feedback for ice-sheet retreat) through slightly enhanced precipitation. However, the precipitation effect is totally overwhelmed by the temperature effect. Our results suggest that the last deglaciation was initiated through increasing summer insolation with CO₂ providing a powerful feedback. Received: 22 February 2000 / Accepted: 17 September 2000     

Flux correction: tests with a simple ocean-atmosphere model

 Flux correction schemes are used in order to suppress the drift of coupled ocean atmosphere models. This technique is tested for a simple box model of the climate system. Two “perfect” models of the ocean and the atmosphere are available. These are coupled to form an ocean-atmosphere model representing the true climate system. This climate system is simulated by a climate model which is also constructed by coupling those two perfect models. This time, however, both models are run first separately as models of the atmosphere and the ocean. In that case, “observations” from the climate system are prescribed at the ocean surface in the uncoupled models. It is assumed that these observations are imperfect. A drift results, when these models are coupled to form an ocean-atmosphere stimulation model. A flux adjustment scheme is implemented to remove this drift. It is argued that the merits and shortcomings of the flux correction technique can be assessed more clearly this way than by coupling imperfect models as is done normally. Sensitivity tests are performed where either radiation parameters are changed or a salt anomaly is implanted. Model parameters are chosen such that the ocean has a thermally direct circulation in the unperturbed climate state. It is found that the flux correction technique is performing satisfactorily as long as the imposed perturbations are small enough so that the ocean circulation does not change its sense. If, however, the model climate is close to the transition to an indirect circulation, then the flux correction technique is unreliable. The predictions of the coupled model with flux correction may deviate substantially from the response of the climate system in that case. Received: 4 December 1995/Accepted: 15 October 1996     

Detection of anthropogenic climate change using an atmospheric GCM

 Atmosphere-only general circulation models are shown to be a useful tool for detecting an anthropogenic effect on climate and understanding recent climate change. Ensembles of atmospheric runs are all forced with the same observed changes in sea surface temperatures and sea-ice extents but differ in terms of the combinations of anthropogenic effects included. Therefore, our approach aims to detect the `immediate' anthropogenic impact on the atmosphere as opposed to that which has arisen via oceanic feedbacks. We have adapted two well-used detection techniques, pattern correlations and fingerprints, and both show that near-decadal changes in the patterns of zonal mean upper air temperature are well simulated, and that it is highly unlikely that the observed changes could be accounted for by sea surface temperature variations and internal variability alone. Furthermore, we show that for zonally averaged upper air temperature, internal `noise' in the atmospheric model is small enough that a signal emerges from the data even on interannual time scales; this would not be possible in a coupled ocean-atmosphere general circulation model. Finally, although anthropogenic forcings have had a significant impact on global mean land surface temperature, we find that their influence on the pattern of local deviations about this mean is so far undetectable. In order to achieve this in the future, as the signal grows, it will also be important that the response of the Northern Hemisphere mid-latitude westerly flow to changing sea surface temperatures is well simulated in climate model detection studies. Received: 3 December 1999 / Accepted: 30 October 2000     

A flexible climate model for use in integrated assessments

 Because of significant uncertainty in the behavior of the climate system, evaluations of the possible impact of an increase in greenhouse gas concentrations in the atmosphere require a large number of long-term climate simulations. Studies of this kind are impossible to carry out with coupled atmosphere ocean general circulation models (AOGCMs) because of their tremendous computer resource requirements. Here we describe a two dimensional (zonally averaged) atmospheric model coupled with a diffusive ocean model developed for use in the integrated framework of the Massachusetts Institute of Technology (MIT) Joint Program on the Science and Policy of Global Change. The 2-D model has been developed from the Goddard Institute for Space Studies (GISS) GCM and includes parametrizations of all the main physical processes. This allows it to reproduce many of the nonlinear interactions occurring in simulations with GCMs. Comparisons of the results of present-day climate simulations with observations show that the model reasonably reproduces the main features of the zonally averaged atmospheric structure and circulation. The model’s sensitivity can be varied by changing the magnitude of an inserted additional cloud feedback. Equilibrium responses of different versions of the 2-D model to an instantaneous doubling of atmospheric CO₂ are compared with results of similar simulations with different AGCMs. It is shown that the additional cloud feedback does not lead to any physically inconsistent results. On the contrary, changes in climate variables such as precipitation and evaporation, and their dependencies on surface warming produced by different versions of the MIT 2-D model are similar to those shown by GCMs. By choosing appropriate values of the deep ocean diffusion coefficients, the transient behavior of different AOGCMs can be matched in simulations with the 2-D model, with a unique choice of diffusion coefficients allowing one to match the performance of a given AOGCM for a variety of transient forcing scenarios. Both surface warming and sea level rise due to thermal expansion of the deep ocean in response to a gradually increasing forcing are reasonably reproduced on time scales of 100–150 y. However a wide range of diffusion coefficients is needed to match the behavior of different AOGCMs. We use results of simulations with the 2-D model to show that the impact on climate change of the implied uncertainty in the rate of heat penetration into the deep ocean is comparable with that of other significant uncertainties. Received: 10 March 1997 / Accepted: 20 October 1997     

Response of the Western European climate to a collapse of the thermohaline circulation

Two ensemble simulations with the ECHAM5/MPI-OM climate model have been investigated for the atmospheric response to a thermohaline circulation (THC) collapse. The model forcing was specified from observations between 1950 and 2000 and it followed a rising greenhouse gases emission scenario from 2001 to 2100. In one ensemble, a THC collapse was induced by adding freshwater in the northern North Atlantic, from 2001 onwards. After about 20 years, an almost stationary response pattern develops, that is, after the THC collapse, global mean temperature rises equally fast in both ensembles with the hosing ensemble displaying a constant offset. The atmospheric response to the freshwater hosing features a strong zonal gradient in the anomalous 2-m air temperature over Western Europe, associated with a strong land–sea contrast. Since Western Europe climate features a strong marine impact due to the prevailing westerlies, the question arises how such a strong land–sea contrast can be maintained. We show that a strong secondary cloud response is set up with increased cloud cover over sea and decreased cloud cover over land. Also, the marine impact on Western European climate decreases, which results from a reduced transport of moist static energy from sea to land. As a result, the change in lapse rate over the cold sea surface temperature (SST) anomalies west of the continent is much larger than over land, dominated by changes in moisture content rather than temperature.     

Impact of bio-physical feedbacks on the tropical climate in coupled and uncoupled GCMs

The bio-physical feedback process between the marine ecosystem and the tropical climate system is investigated using both an ocean circulation model and a fully-coupled ocean–atmosphere circulation model, which interact with a biogeochemical model. We found that the presence of chlorophyll can have significant impact on the characteristics of the El Niño-Southern Oscillation (ENSO), including its amplitude and asymmetry, as well as on the mean state. That is, chlorophyll generally increases mean sea surface temperature (SST) due to the direct biological heating. However, SST in the eastern equatorial Pacific decreases due to the stronger indirect dynamical response to the biological effects outweighing the direct thermal response. It is demonstrated that this biologically-induced SST cooling is intensified and conveyed to other tropical-ocean basins when atmosphere–ocean coupling is taken into account. It is also found that the presence of chlorophyll affects the magnitude of ENSO by two different mechanisms; one is an amplifying effect by the mean chlorophyll, which is associated with shoaling of the mean thermocline depth, and the other is a damping effect derived from the interactively-varying chlorophyll coupled with the physical model. The atmosphere–ocean coupling reduces the biologically-induced ENSO amplifying effect through the weakening of atmospheric feedback. Lastly, there is also a biological impact on ENSO which enhances the positive skewness. This skewness change is presumably caused by the phase dependency of thermocline feedback which affects the ENSO magnitude.     

Modelling climate change impacts on maize growth and development in the Czech Republic

Summary The crop growth model CERES-Maize is used to estimate the direct (through enhanced fertilisation effect of ambient CO₂) and indirect (through changed climate conditions) effects of increased concentration of atmospheric CO₂ on maize yields. The analysis is based on multi-year crop model simulations run with daily weather series obtained alternatively by a direct modification of observed weather series and by a stochastic weather generator. The crop model is run in two settings: stressed yields are simulated in water and nutrient limited conditions, potential yields in water and nutrient unlimited conditions. The climate change scenario was constructed using the output from the ECHAM3/T42 model (temperature), regression relationships between temperature and solar radiation, and an expert judgement (precipitation). Results: (i) After omitting the two most extreme misfits, the standard error between the observed and modelled yields is 11%. (ii) The direct effect of doubled CO₂: The stressed yields would increase by 36–41% in the present climate and by 61–66% in the 2 × CO₂ climate. The potential yields would increase only by 9–10% as the improved water use efficiency does not apply. (iii) The indirect effect of doubled CO₂: The stressed yields would decrease by 27–29% (14–16%) at present (doubled) ambient CO₂ concentration. The increased temperature shortens the phenological phases and does not allow for the optimal development of the crop. The simultaneous decrease of precipitation and increase of temperature and solar radiation deepen the water stress, thereby reducing the yields. The reduction of the potential yields is significantly smaller as the effect of the increased water stress does not apply. (iv) If both direct and indirect effects of doubled CO₂ are considered, the stressed yields should increase by 17–18%, and the potential yields by 5–14%. (v) The decrease of the stressed yields due to the indirect effect may be reduced by applying earlier planting dates. Received March 9, 2001 Revised September 25, 2001     

The effect of atmospheric CO2 and ice sheet topography on LGM climate

The role of reduced atmospheric CO₂ concentration and ice sheet topography plus its associated land albedo on the LGM climate is investigated using a coupled atmosphere-ocean-sea ice climate system model. The surface cooling induced by the reduced CO₂ concentration is larger than that by the ice sheet topography plus other factors by about 30% for the surface air temperature and by about 100% for the sea surface temperature. A large inter-hemispheric asymmetry in surface cooling with a larger cooling in the Northern Hemisphere is found for both cases. This asymmetric inter-hemispheric temperature response is consistent in the ice sheet topography case with earlier studies using an atmospheric model coupled with a mixed-layer ocean representation, but contrasts with these results in the reduced CO₂ case. The incorporation of ocean dynamics presumably leads to a larger snow and sea ice feedback as a result of the reduction in northward ocean heat transport, mainly as a consequence of the decrease in the North Atlantic overturning circulation by the substantial freshening of the North Atlantic convection regions. A reversed case is found in the Southern Ocean. Overall, the reduction in atmospheric CO₂ concentration accounts for about 60% of the total LGM climate change.     

A comparison of PMIP2 model simulations and the MARGO proxy reconstruction for tropical sea surface temperatures at last glacial maximum

Results from multiple model simulations are used to understand the tropical sea surface temperature (SST) response to the reduced greenhouse gas concentrations and large continental ice sheets of the last glacial maximum (LGM). We present LGM simulations from the Paleoclimate Modelling Intercomparison Project, Phase 2 (PMIP2) and compare these simulations to proxy data collated and harmonized within the Multiproxy Approach for the Reconstruction of the Glacial Ocean Surface Project (MARGO). Five atmosphere–ocean coupled climate models (AOGCMs) and one coupled model of intermediate complexity have PMIP2 ocean results available for LGM. The models give a range of tropical (defined for this paper as 15°S–15°N) SST cooling of 1.0–2.4°C, comparable to the MARGO estimate of annual cooling of 1.7 ± 1°C. The models simulate greater SST cooling in the tropical Atlantic than tropical Pacific, but interbasin and intrabasin variations of cooling are much smaller than those found in the MARGO reconstruction. The simulated tropical coolings are relatively insensitive to season, a feature also present in the MARGO transferred-based estimates calculated from planktonic foraminiferal assemblages for the Indian and Pacific Oceans. These assemblages indicate seasonality in cooling in the Atlantic basin, with greater cooling in northern summer than northern winter, not captured by the model simulations. Biases in the simulations of the tropical upwelling and thermocline found in the preindustrial control simulations remain for the LGM simulations and are partly responsible for the more homogeneous spatial and temporal LGM tropical cooling simulated by the models. The PMIP2 LGM simulations give estimates for the climate sensitivity parameter of 0.67°–0.83°C per Wm⁻², which translates to equilibrium climate sensitivity for doubling of atmospheric CO₂ of 2.6–3.1°C.     

Impact of Greenland and Antarctic ice sheet interactions on climate sensitivity

We use the Earth system model of intermediate complexity LOVECLIM to show the effect of coupling interactive ice sheets on the climate sensitivity of the model on a millennial time scale. We compare the response to a 2×CO₂ warming scenario between fully coupled model versions including interactive Greenland and Antarctic ice sheet models and model versions with fixed ice sheets. For this purpose an ensemble of different parameter sets have been defined for LOVECLIM, covering a wide range of the model??s sensitivity to greenhouse warming, while still simulating the present-day climate and the climate evolution over the last millennium within observational uncertainties. Additional freshwater fluxes from the melting ice sheets have a mitigating effect on the model??s temperature response, leading to generally lower climate sensitivities of the fully coupled model versions. The mitigation is effectuated by changes in heat exchange within the ocean and at the sea?Cair interface, driven by freshening of the surface ocean and amplified by sea?Cice-related feedbacks. The strength of the effect depends on the response of the ice sheets to the warming and on the model??s climate sensitivity itself. The effect is relatively strong in model versions with higher climate sensitivity due to the relatively large polar amplification of LOVECLIM. With the ensemble approach in this study we cover a wide range of possible model responses.     

Simulation of the Indian monsoon using the RegCM3–ROMS regional coupled model

A regional coupled atmosphere–ocean model was developed to study the role of air–sea interactions in the simulation of the Indian summer monsoon. The coupled model includes the regional climate model (RegCM3) as atmospheric component and the regional ocean modeling system (ROMS) as oceanic component. The two-way coupled model system exchanges sea surface temperature (SST) from the ocean to the atmospheric model and surface wind stress and energy fluxes from the atmosphere to the ocean model. The coupled model is run for four years 1997, 1998, 2002 and 2003 and the results are compared with observations and atmosphere-only model runs employing Reynolds SSTs as lower boundary condition. It is found that the coupled model captures the main features of the Indian monsoon and simulates a substantially more realistic spatial and temporal distribution of monsoon rainfall compared to the uncoupled atmosphere-only model. The intraseasonal oscillations are also better simulated in the coupled model compared to the atmosphere-only model. These improvements are due to a better representation of the feedbacks between the SST and convection and highlight the importance of air–sea coupling in the simulation of the Indian monsoon.     

Relative roles of climate sensitivity and forcing in defining the ocean circulation response to climate change

The response of the ocean’s meridional overturning circulation (MOC) to increased greenhouse gas forcing is examined using a coupled model of intermediate complexity, including a dynamic 3-D ocean subcomponent. Parameters are the increase in CO₂ forcing (with stabilization after a specified time interval) and the model’s climate sensitivity. In this model, the cessation of deep sinking in the north “Atlantic” (hereinafter, a “collapse”), as indicated by changes in the MOC, behaves like a simple bifurcation. The final surface air temperature (SAT) change, which is closely predicted by the product of the radiative forcing and the climate sensitivity, determines whether a collapse occurs. The initial transient response in SAT is largely a function of the forcing increase, with higher sensitivity runs exhibiting delayed behavior; accordingly, high CO₂-low sensitivity scenarios can be assessed as a recovering or collapsing circulation shortly after stabilization, whereas low CO₂-high sensitivity scenarios require several hundred additional years to make such a determination. We also systemically examine how the rate of forcing, for a given CO₂ stabilization, affects the ocean response. In contrast with previous studies based on results using simpler ocean models, we find that except for a narrow range of marginally stable to marginally unstable scenarios, the forcing rate has little impact on whether the run collapses or recovers. In this narrow range, however, forcing increases on a time scale of slow ocean advective processes results in weaker declines in overturning strength and can permit a run to recover that would otherwise collapse.     

Towards the detection and attribution of an anthropogenic effect on climate

It has been hypothesized recently that regional-scale cooling caused by anthropogenic sulfate aerosols may be partially obscuring a warming signal associated with changes in greenhouse gas concentrations. Here we use results from model experiments in which sulfate and carbon dioxide have been varied individually and in combination in order to test this hypothesis. We use centered [R (t)] and uncentered [C (t)] pattern similarity statistics to compare observed time-evolving surface temperature change patterns with the model-predicted equilibrium signal patterns. We show that in most cases, the C (t) statistic reduces to a measure of observed global-mean temperature changes, and is of limited use in attributing observed climate changes to a specific causal mechanism. We therefore focus on R (t), which is a more useful statistic for discriminating between forcing mechanisms with different pattern signatures but similar rates of global mean change. Our results indicate that over the last 50 years, the summer (JJA) and fall (SON) observed patterns of near-surface temperature change show increasing similarity to the model-simulated response to combined sulfate aerosol/CO₂ forcing. At least some of this increasing spatial congruence occurs in areas where the real world has cooled. To assess the significance of the most recent trends in R (t) and C (t), we use data from multi-century control integrations performed with two different coupled atmosphere-ocean models, which provide information on the statistical behavior of 'unforced' trends in the pattern correlation statistics. For the combined sulfate aerosol/CO₂ experiment, the 50-year R (t) trends for the JJA and SON signals are highly significant. Results are robust in that they do not depend on the choice of control run used to estimate natural variability noise properties. The R (t) trends for the CO₂-only signal are not significant in any season. C (t) trends for signals from both the CO₂-only and combined forcing experiments are highly significant in all seasons and for all trend lengths (except for trends over the last 10 years), indicating large global-mean changes relative to the two natural variability estimates used here. The caveats regarding the signals and natural variability noise which form the basis of this study are numerous. Nevertheless, we have provided first evidence that both the largest-scale (global-mean) and smaller-scale (spatial anomalies about the global mean) components of a combined CO₂/anthropogenic sulfate aerosol signal are identifiable in the observed near-surface air temperature data. If the coupled-model noise estimates used here are realistic, we can be highly confident that the anthropogenic signal that we have identified is distinctly different from internally generated natural variability noise. The fact that we have been able to detect the detailed spatial signature in response to combined CO₂ and sulfate aerosol forcing, but not in response to CO₂ forcing alone, suggests that some of the regional-scale background noise (against which we were trying to detect a CO₂-only signal) is in fact part of the signal of a sulfate aerosol effect on climate. The large effect of sulfate aerosols found in this study demonstrates the importance of their inclusion in experiments designed to simulate past and future climate change. Received: 10 November 1994 / Accepted: 19 July 1995     

On the cold start problem in transient simulations with coupled atmosphere-ocean models

Finite computer resources force compromises in the design of transient numerical experiments with coupled atmosphere-ocean general circulation models which, in the case of global warming simulations, normally preclude a full integration from the undisturbed pre-industrial state. The start of the integration at a later time from a climate state which, in contrast to the true climate, is initially in equilibrium then induces a cold start error. Using linear response theory a general expression for the cold start error is derived. The theory is applied to the Hamburg CO₂ scenario simulations. An attempt to estimate the global-mean-temperature response function of the coupled model from the response of the model to a CO₂ doubling was unsuccessful because of the non-linearity of the system. However, an alternative derivation, based on the transient simulation itself, yielded a cold start error which explained the initial retardation of the Hamburg global warming curve relative to the IPCC results obtained with a simple box-diffusion-upwelling model. In the case of the sea level the behaviour of the model is apparently more linear. The cold start error estimations based on a CO₂ doubling experiment and on an experiment with gradually increasing CO₂ (scenario A) are very similar and explain about two thirds of the coupled model retardation relative to the IPCC results.     

Towards quantifying uncertainty in transient climate change

Ensembles of coupled atmosphere–ocean global circulation model simulations are required to make probabilistic predictions of future climate change. “Perturbed physics” ensembles provide a new approach in which modelling uncertainties are sampled systematically by perturbing uncertain parameters. The aim is to provide a basis for probabilistic predictions in which the impact of prior assumptions and observational constraints can be clearly distinguished. Here we report on the first perturbed physics coupled atmosphere–ocean model ensemble in which poorly constrained atmosphere, land and sea-ice component parameters are varied in the third version of the Hadley Centre model (the variation of ocean parameters will be the subject of future study). Flux adjustments are employed, both to reduce regional sea surface temperature (SST) and salinity biases and also to admit the use of combinations of model parameter values which give non-zero values for the global radiation balance. This improves the extent to which the ensemble provides a credible basis for the quantification of uncertainties in climate change, especially at a regional level. However, this particular implementation of flux-adjustments leads to a weakening of the Atlantic overturning circulation, resulting in the development of biases in SST and sea ice in the North Atlantic and Arctic Oceans. Nevertheless, model versions are produced which are of similar quality to the unperturbed and un-flux-adjusted version. The ensemble is used to simulate pre-industrial conditions and a simple scenario of a 1% per year compounded increase in CO₂. The range of transient climate response (the 20 year averaged global warming at the time of CO₂ doubling) is 1.5–2.6°C, similar to that found in multi-model studies. Measures of global and large scale climate change from the coupled models show simple relationships with associated measures computed from atmosphere-mixed-layer-ocean climate change experiments, suggesting that recent advances in computing the probability density function of climate change under equilibrium conditions using the perturbed physics approach may be extended to the transient case.